ResearchDatasetsSkyworkFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPreference AlignmentSkywork80Win Rate (W)24Language ModelingSkywork3.25Finance Score8Direct Preference OptimizationSkywork AlpacaEval 2.020.56LCWR7Human EvaluationSkywork (test)1,610.1Elo Rating5